- Email: [email protected]
Changes in human coronary sinus blood flow and myocardial metabolism induced by ventricular fibrillation and defibrillation
Changes in Human Coronary Sinus Blood Flow and Myocardial Metabolism Induced by Ventricular Fibrillation and Defibrillation Mikael Runsi6, MD, Lennart Bergfeldt, MD, PhD, M~rten Rosenqvist, MD, PhD, Anders Owall, MD, PhD, and Lennart Jorfeldt, MD, PhD Background: During implantation of cardioverter-defibrillators, repeated inductions of ventricular fibrillation and defibrillation are performed. Little is known about the myocardial metabolism associated with ventricular fibrillation and defibrillation in humans. Methods: Sixteen patients scheduled for transvenous cardioverter-defibrillator implantation were included in the study. In 10 of the patients, blood samples were taken simultaneously in the coronary sinus and radial artery and analyzed for PO2, PCO2, standard bicarbonate, pH, lactate, alanine, glucose, and glycerol. Oxygen saturation, base excess, and oxygen content were calculated. The patients were studied before, shortly after, and 2 and 5 minutes after successful defibrillation. In six of the patients, coronary sinus blood flow was registered continuously. Results: The coronary sinus blood flow declined from a basal value of 93 -+ 16 mL/min to 35 -+ 6 mL/min 14 -+ 2 seconds after induction of ventricular fibrillation. Following
termination of ventricular fibrillation, coronary sinus blood flow increased to a peak value of 227 -+ 75 mL/min. Oxygen saturation, PO2, and oxygen content in the coronary sinus increased by approximately 25% shortly after each episode of ventricular fibrillation and defibrillation. The coronary sinus lactate increased and the arterio-coronary sinus lactate difference decreased shortly after each of the four episodes, but was normalized within 2 minutes. Conclusions: Repeated threshold tests during defibrillator implantation did not cause any long-lasting or cumulative metabolic effects, indicating that the described technique, with a 5-minute recovery period in between episodes, is safe as regards myocardial metabolism.
INCE THE introduction of the implantable cardioverterdefibrillator (ICD) into clinical practice, an increasing number of patients with malignant ventricular arrhythmias are now offered symptomatic treatment of an otherwise lifethreatening condition. 1 The implantation procedure, although less traumatic today, produces a temporary depression of myocardial function with the risk of further impairment in myocardial performance after repeated episodes of ventricular fibrillation and defibrillation.2-4 During the implantation procedure of an ICD, the heart is exposed to several myocardial depressant factors, in which the short-lasting ventricular fibrillation and the direct current (DC) shock delivered to the heart constitute the most probable sources of myocardM injury. Ventricular fibrillation and its effects on myocardial metabolism and the neurohumoral response have been studied ia several experimental settings.5-s So far, no study of the metabolic effects has been conducted in humans during implantation of an ICD. In animal experiments, the metabolic effects of transthoracic DC countershocks showed a large increase in coronary sinus lactate level after five defibrillator shocks. T M Other studies have shown myocardial injury indicated by the release of troponin-T and other markers of myocardial injury.l:'- The aim of this study was to evaluate the metabolic effects of repeated short episodes of ventricular fibrillation and internal defibrillation in relation to coronary artery blood flow.
6 patients, continuous coronary sinus blood flow was recorded. Clinical data for the two groups are shown in Table 1. After an overnight fast, the patients were premedicated with morphine, 0.05 to 0.1 mg/kg, and scopolamine, 2 to 4 Bg/kg.Anesthesia was induced using thiopental sodium, 3 to 5 mg/kg, and neuromuscular blockade was achieved using vecttronium, 0.1 mg/kg. Anesthesia was maintained with isoflurane in oxygen/nitrous oxide, and the patients were ventilated to achieve normocapnia. Before induction of ventricular fibrillation, the FIO2 was increased to 100%. In all 16 patients, endocardial leads (Medtronic Inc, Minneapolis, MN, or Cardiac Pacemakers Inc, St Paul, MN) were introduced through either the left cephalic or subclavian vein and combined with a subcutaneous patch, when necessary. The transvenous leads were positioned under fluoroscopic guidance, with one large coil electrode in the apex of the right ventricle and, when the Medtronic system was used, one thinner electrode was positioned in the junction of the superior vena cava and the right atrium. After an appropriate position was found, the electrodes were connected to the test equipment, and the pacing threshold and impedance as well as the slew rate and R-wave amplitude were assessed. When these values fulfilled the implantation criteria, ventricular fibrillation was induced, either by alternating current (AC) or by T-wave shock, and defibrillated with a maximal energy of 20 joules. After at least three successful tests, the electrodes were connected to the generator, which was implanted in an abdominal pocket, after which a final test with 34 joules was performed. After the induction of anesthesia, but before surgery, a thermodilution catheter (Webster Laboratories, Altadena, CA) was introduced into the coronary sinus, guided by fluoroscopy for flow measurements and blood
S
MATERIAL AND METHODS This study was approved by the Ethics Committee of the Karolinska Hospital, and the patients gave their informed consent to participate. Sixteen patients with malignant ventricular arrhythmias were studied during the implantation procedure of an ICD with regard to the effects of repeated episodes of ventricular fibrillation and defibrillation. For technical reasons, it was not possible to perform continuous measurements of coronary sinus blood flow and blood sampling in the same group of patients. In group A, consisting of 10 patients, coronary sinus blood flow and absolute and arterio-coronary sinus differences of energy substrates were intermittently assessed. In group B, consisting of
Copyright© 1998by W.B, Saunders Company KEY WORDS: ventricular fibrillation, defibrillation, implantable cardioverter-defibrillator, myocardial metabolism
From the Departments of Cardiothoracic Surgery, Cardiology, Cardiothoracic Anesthesiology, and Clinical Physiology, Karolinska Institute at Karolinska Hospital, Stockholm, Sweden. This study was supported by grants from the Swedish Heart and Lung Foundation and the Swedish Medical Research Council, Project No. 04139. Address reprint requests to Mikael Runsi5, MD, Department of Cardiothoracic Surgery, IFX Karolinska Hospital, S-171 76 Stockholm, Sweden. Copyright © 1998 by W.B. Saunders Company 1053-0770/98/1201-0008503.00/0
Journal of Cardiothoracic and Vascular Anesthesia, Vo112, No 1 (February),1998:pp 45-50
45
46
RUNSIO ET AL
Table 1. Patient Characteristics Variable/Group
Group A n - 10
Male (%)
80
Age (yr)
58 _+ 12
Failed drugs (no)
2.8 _+ 2
Group B n- 6 100 68 _+ 4 2.2 -- 1.2
Sotalol at surgery
7
4
Amiodarone at surgery
0
0
Previous heart surgery Heart disease
3
1
CAD DCM
7 1
5 0
NSHD
2
I
VT VF
4 1
4 1
VT/VF
4
1
SD
1
0
Arrhythmia
W EF Duration of operation (min) Inductions (no) VF duration (sec)
110 _+ 70 0.37 _+ 0.15
133 _+ 34 0.44 _+ 0.17
150 -- 30 6 _+ 1 16 -- 5
140 +_ 30 5 _+ 1 17 _+ 1
Abbreviations: CAD, coronary artery disease; DCM, dilated cardiomyopathy; EF, ejection fraction; NSHD, no structural heart disease; SD, sudden death; VF, ventricular fibrillation; VT, ventricular tachycardia; W, work test.
sampling. A catheter for blood sampling and blood pressure measurement was placed in the radial artery. The coronary sinus blood flow (mL/min) was determined by the retrograde thermodilution technique using a continuous infusion of saline over 20 seconds at a rate of 16.4 m L / m i n . 5 In group A patients, flow measurements were performed before the first induction of ventricular fibrillation and during the period 2 to 5 minutes after each episode of ventricular fibrillation and defibrillation, when stable hemodynamic conditions were re-established. In group B patients, the coronary sinus blood flow and the arterial blood pressure were measured continuously, and the mean value was recorded every 4 seconds, from 40 seconds before to 80 seconds after each of three episodes of ventricular fibrillation and defibrillation. The repayment/debt volume ratio was calculated by estimating the flow difference below (debt) and above (repayment) the baseline value. Blood samples were taken before induction of ventricular fibrillation, shortly after successful defibrillation, and 2 and 5 minutes after defibrillation. The 5-minute sample also served as the prefibrillation sample for the following episode, and, in each patient, 13 samples were taken from the coronary sinus and radial artery, respectively. Blood samples were obtained between 8 and 20 seconds and between 20 and 40 seconds after defibrillation for the determination of blood gas and metabolic variables, respectively. Each set of blood samples was taken in two portions: the first was drawn into a heparinized syringe and immediately placed on ice before it was analyzed for blood gas variables and calculations of base excess; oxygen saturation and oxygen were performed using standard formulas (Blood gas analyzer, model IL1312, BGM Instrumentation Laboratory Inc, Milan, Italy). The second portion, drawn directly after the blood gas portion, was immediately precipitated with ice-cold perchloric acid. After centrifugation, the protein-free extract was deep-frozen (-80°C) while awaiting analysis. The analyses were performed by microfluorimetry: D-glucose by a modification for fluorimetry of the hexokinase method, and lactate, glycerol, 3-hydroxybutyrate, and alanine as previously described.13-15 The data were analyzed in accordance with the analysis of variance (ANOVA) for repeated measures. Repeating factors were episodes with four levels and time with three levels. Because of within-subject
correlations among the repeated measurements, the p values have been adjusted according to Greenhouse-Geisser. When sphericity assumption in the repeated measures design did not hold, a multivariate analysis of variance was performed. When the overall F-ratio in the analysis of variance was significant at a 5% level, contrast (trends) between the repeated measures was calculated. The distribution for some of the variables was skewed. The data have been logtransformed to meet the requirements for an adequate ANOVA. 16 Statistical significance was defined asp < 0.05. The data are presented as mean + standard error of the mean. RESULTS
The patients in group A had a total of 6 + 1 episodes of ventricular fibrillation and defibrillations, and group B patients had 5 ± 1 episodes. The ventricular fibrillation lasted for 16 ± 5 seconds and 17 --- 1 seconds before defibrillation in groups A and B, respectively. Group A patients received 129 ± 12 joules/patient, and group B patients 149 --- 23 joules/patient. The coronary sinus blood flow in group A was 73 ± 15 mL/min at baseline and 69 ± 5 mL/min, 79 + 12 mL/min, 62 8 mL/min, and 63 ± 11 mL/min (NS), respectively, in the period 2 to 5 minutes after each of the four episodes of ventricular fibrillation and defibrillation. In group B, basal coronary sinus blood flow before the first episode of fibrillation was 93 ± 16 mL/min (Fig 1). During ventricular fibrillation, the coronary sinus blood flow decreased to 35 ± 6 mL/min (p < 0.001). Immediately after defibrillation, there was a marked increase in coronary sinus blood flow, reaching a maximum of 227 ± 75 mL/min (p < 0.001) within 20 ± 5 seconds after defibrillation. After a further 30 to 40 seconds, the coronary sinus blood flow had returned to prefibrillation levels (Fig 1). The repayment/debt volume ratio was 2.7 ± 0.4, 2.9 ± 0.3, and 3.1 ± 0.4 in three consecutive episodes of ventricular fibrillation and defibrillation, respectively. Mean arterial blood pressure decreased to 29 ± 3 mmHg during fibrillation from a basal value of 72 ± 2 mmHg (p < 0.001). It returned to the prefibrillation level of 71 ± 3 mmHg within 30 seconds following defibrillation (Fig 1). Arterial blood PO2 increased from 283 ± 34 to 417 --- 19 m m H g (p < 0.001) during the study period. Coronary sinus PO2 was stable in the prefibrillation state, but increased significantly shortly after each episode of ventricular fibrillation from a baseline level of 24 ± 1 to 28 ± 2 mmHg ( p < 0.001), 31 - 2 mmHg ( p < 0.001),29 ± 2 mmHg ( p < 0.001), and 31 ± 2 mmHg (p < 0.001), respectively. Within 2 minutes, it had returned to the prefibrillation level. The arterial oxygen saturation was 100% during the study period. The coronary sinus oxygen saturation increased significantly after each of the episodes of ventricular fibrillation and defibrillation. Before the first induction of ventricular fibrillation, the oxygen saturation in coronary sinus blood was 42% ± 2%, increasing to 50% ± 4.5% ( p < 0.001), 58% ± 3% ( p < 0 . 0 0 1 ) , 55% ± 3% ( p < 0 . 0 0 1 ) , and 56% - 5% (p < 0.001) respectively, shortly after the four episodes of ventricular fibrillation and defibrillation. The oxygen content in the coronary sinus increased significantly (p < 0.001) shortly after each of the four episodes of ventricular fibrillation and defibrillation (Fig 2). No significant changes were seen in arterial and coronary sinus PCO2, pH, standard bicarbonate, or base excess during the procedure.
MYOCARDIAL METABOLISM AND DEFIBRILLATION
47
250
80 T]TLLLLU_U_LLLLM.LLLL~
7O
T
T
[
~
_TTrTTTTTTT[TTTTTTIT
TI'T~TT
6o so
.
150
40 "~
30
'c
20
Fig 1. Mean arterial blood pressure = A and coronary sinus blood flow = ~ during three consecutive episodes of ventricular fibrillation and defibrillation in patients in group B. Mean -+ SEM.
100
I 0
r
,
20 40 60 110 100120140 0
,
,
20 40 60 80 100 120140 0
,
~
~
j
,
,
,
8
I0
20 40 60 80 100120140
Sec
The hemoglobin value was stable in both groups (luring the procedure. The mean value for group A patients was 144 4 g/L. The arterial lactate concentration was stable during the whole procedure (0.8 + 0.07 to 0.9 -+ 0.08 mmol/L) (NS). The coronary sinus lactate was stable during the prefibrillation periods. It was 0.6 +- 0.06 mmol/L at baseline and 0.7 + 0.09 mmol/L (p < 0.05), 0.7 + 0.08 mmol/L (p < 0.05) 0.7 _+ 0.06 mmol/L (p < 0.05), and 0.8 + 0.08 mmol/L (p < 0.05) after each episode of fibrillation, respectively. The arteriocoronary sinus lactate difference decreased significantly (p < 0.05) 20 to 40 seconds after each episode of ventricular fibrillation and defibrillation, but returned to baseline values 2 minutes after restoration of regular heart rhythm (Fig 3). There was a net release of lactate in four patients. There were no significant changes in alanine, glycerol, or glucose during the study period. DISCUSSION The implantation procedure of an ICD offers a unique opportunity to study the immediate effects of ventricular fibrillation and defibrillation on myocardial hemoJynamics,
neurohumoral response, and metabolism in a closed-chest model. Applying this concept, it was shown that short episodes of electrically induced ventricular fibrillation cause an immediate postdefibrillation myocardial hyperemia with a concomitant increase in coronary sinus oxygen content. These changes were associated with an increase in coronary sinus lactate and a decreased arteriocoronary sinus lactate difference. However, the observed effects were of short duration, and no signs of a cumulative metabolic derangement were observed. In the present study, the mean arterial blood pressures were stable throughout the procedure, apart from during and 40 seconds after induction of ventricular fibrillation, indicating that the hemodynamics were reasonably stable in the basic condition. In the same model, a progressive diastolic dysfunction without any concomitant deterioration in systolic function has been observed, which is in accordance with other studies. 3,4,17 Despite the reduced driving pressure during fibrillation, there was a persistent, although decreased, myocardial perfusion. This reduction is in accordance with results from previous studies. 4,7-8 After defibrillation, a marked hyperemia was observed. Reactive hyperemia is a well-kalown physiologic reac0.30
14.0
0.25
12.0
t
10,0
E x O
t
~
0.20
8,0
0.15
6.0 4.0
0.10 2.0
J
0.05
0.0 BI
A0
Episode 1
A2
B2
A0
Episode 2
A2
B3
A0
Episode 3
A2
B4
AC
A2
A5
Episode 4
Fig 2. Oxygen content in coronary sinus during four consecutive episodes of ventricular fibrillation and defibrillation in group A patients. Mean + SEM. Abbreviations: B1-B4, before eph,;ode 1 to 4; A0, shortly after defibrillation; A2, 2 minutes after defibrillation; and A5, 5 minutes after defibrillation.
0.00
A
A2
A5
Fig 3. Arteriocoronary sinus lactate difference during four consecutive episodes of ventricular fibrillation and defibrillation in group A patients. Mean -+ SEM. Abbreviations: B, before; A, after defibrillation; A2, 2 minutes after defibrillation; A5, 5 minutes after defibrillation; r~, Episode 1; II, Episode 2; [], Episode 3; D , Episode 4.
48
A
RUNSIO ET AL
B
6oo
600 o
.E
O
E •~. 500
500
O
O
0
o 400
o
8
o
o
.El
400
o
o
o
c
'~ 200
~o
o
0
o
Oo
° i 0 -°
oll
0 8e u 8°oO ~e ~o o g ^ l l
O0
100
2 o
p-II-dl'-1F8
L)
0
I
8~
8
o
o
i
I
I
I
20
40
60
80
Mean arterial blood pressure
(ram Hg}
0
8
o 300
'~ 200
0
O O
300
O
o o°
lo0
°o o - e ~ -
@
O
°i
o 0
0
i
I
I
i
20
40
60
80
Mean arterial blood pressure (mm Hg}
Fig 4. (A) Mean arterial blood pressure versus coronary sinus blood flow before induction of, and during, ventricular fibrillation in group B patients. The line indicates correlation using the least square method. Arrow indicates direction of pressure and coronary sinus blood flow development. (B) Mean arterial blood pressure versus coronary sinus blood flow immediately after defibrillation and during reactive hyperemia in group B patients. Interrupted line indicates correlation using the least square method. Arrow indicates direction of pressure and coronary sinus blood flow development.
tion after reperfusion after regional impairment of arterial blood flow.18,19This phenomenon is considered to be caused by both physical and metabolic factors, and has been observed in several organs, including the heart. 2°-22Figure 4 is a scatterplot of coronary sinus blood flow versus mean arterial blood pressure during fibrillation with successively decreasing mean arterial blood pressure (Fig 4A) and during reperfusion (Fig 4B). During fibrillation, coronary sinus blood flow decreased approximately in parallel with the mean arterial blood pressure. This could indicate that the coronary vascular resistance was essentially unchanged. During reperfusion, approximately the same relationship occurred, initially at low mean arterial blood pressures. 22 After this, however, the coronary sinus blood flow was higher than predicted from the mean arterial blood pressure, indicating a decrease in the coronary vascular resistance probably because of the release of vasoactive agents, a3 Adenosine release may be one plausible mechanism responsible for the postfibrillation/defibrillationhyperemia seen in these patients. The release of adenosine, in association with the reactive hyperemic response seen after occlusion of a coronary artery or after ventricular fibrillation, has been observed by others. 6,23In the present study, there was a hypoperfusion of 17 -+ 1 second, approximately the same duration as in the study by Olsson et al,2° with a considerable hyperemic response shortly after defibrillation. Other investigators have studied the volume repayment/debt ratio and found that the repayment volume always exceeds the debt volume, which is in accordance with the authors' current results. 18,z.26 The repayment/debt ratio in the present study was 3:1, as has been described by others in occlusion studies of the same duration. This overperfusion could indicate a temporary uncoupling between oxygen supply and demand. The very high efficiency of the compensatory mechanisms involved in the regulation of myocardial perfusion is further illustrated in a study by Schwartz et al,27 in which occlusion episodes as brief as 200 milliseconds were enough to create an immediate hyperemic response during the next heartbeat.
In the present study, the coronary sinus oxygen saturation increased by 25% shortly after the fibrillation/defibrillation episodes. Coronary sinus oxygen saturation could not be measured continuously during ventricular fibrillation. Therefore, the myocardial oxygen debt and repayment could not be calculated. Considering, however, that the repayment/debt ratio of coronary sinus blood flow was high (3:1) when compared with the moderate increase in the coronary sinus oxygen saturation, it seems as if an oxygen debt was not only repaid, but, to some extent, overcompensated for. This reaction is probably a consequence of the hyperemia alone. An alternative cause of the increase in coronary sinus oxygen content could be a depressed mitochondrial oxidative phosphorylation in response to the delivered DC countershocks, as described by Trouton et 1tl. 9-11 The response to the combined arrhythmic and electrical trauma is Complex, and several interacting factors probably contribute to the observed increase in coronary sinus oxygen content.29 Coronary sinus lactate increased shortly after defibrillation, with a concomitant reduction in the arteriocoronary sinus lactate difference seen particularly after the fourth episode. However, measurements in association with ventricular fibrillation and defibrillation are generally made under non-steadystate conditions, complicating the interpretation of the results. 28 To reduce this problem, a comparison between arteriocoronary sinus differences of lactate with that of oxygen was performed (Fig 5). Provided that the compared substances have a similar distribution volume and transit time through the tissue, a similar response to the intervention could be predicted. 28 If this could be assumed for lactate and oxygen, it further supports the ideas put forward in this study, indicating a decreased lactate uptake seen shortly after the episodes of DC cardioversion, most prominent after episodes 1 and 4 (Fig 5). An increase in coronary sinus lactate during these non-steady-state conditions may indicate ischemia, although a net release of lactate is the only reliable sign of ischemia. Coronary sinus lactate increased after veutricular fibrillation, concomitant with a pronounced
MYOCARDIAL METABOLISM AND DEFIBRILLATION
49
0.06 0.05
ttttt t
0.04 0.03 0.02 0.01 0.00 -0.01 -0.02
B1
A0
Episode 1
A2
B2
AO
Episode 2
A2
B3
A0
Episode 3
A2
B4
A0
A2
A5
Episode 4
Fig 5. The arteriocoronary sinus lactate versus oxygen difference in group A patients. Mean _+ SEM. Abbreviations: B1-B4, before episode 1 to 4; A0, immediately after defibrillation; A2, 2 minutes after defibrillation; and A5, 5 minutes after defibrillation.
increase in coronary sinus blood flow, but without corresponding increase in oxygen demand. Lactate concentration may have increased because of ischemia, but may also have been altered by the marked increase in myocardial blood flow and the corresponding increase in substrate (lactate) delivery. Determination of lactate during reactive hyperemia may, therefore, be a blunt tool for the detection of myocardial ischemia. The coronary artery pressure was severely depressed for 20 to 30 seconds, which could be one possible cause of the effects on lactate metabolism seen in these patients. This is in accordance with studies by Olsson et al,u,25 in which a hypoperfusion of 6 seconds was enough for lactate production to appear. As described by Trouton et al, 9-n the electrical energy delivered in sinus rhythm could cause a significant lactate release. This seemed to be a dose-response effect because this effect was caused by five shocks, but not by two. Thus, both the low coronary artery pressure cau:;ed by the arrhythmia and the defibrillation energy could be of importance for the changes seen in myocardial metabolism, as indicated by the decreased arteriocoronary sinus lactate difference: after ICD threshold tests. This study was performed on a limited number of patients
with various underlying diseases and ventricular arrhythmias, which reflects the clinical situation. However, the data are in accordance with previous animal studies. 9-11 Another limitation is that the experimental setup did not allow simultaneous measurements of blood flow and blood sampling from the coronary sinus, which makes all attempts to calculate ~:he absolute release or uptake of different substances precarious. A further limitation is that the sampling of metabolites and blood gas components in coronary sinus would only give a net result of the release of the measured substance. If regional productmn and release of a substance, as described for lactate, could be suspected, this method has obvious disadvantages, raising the possibility of failing to detect a significant release of clinical importance. 3°-32 The measurements were made shortly after defibrillation, during non-steady-state conditions, further cemplicating the analysis of the results as previously discussed. The decrease in coronary sinus blood flow after induction of ventricular fibrillation is probably an effect of decreased driving pressure over the coronary vascular bed, whereas the postdefibrillation hyperemia seen in all patients after defibrillation might be a result of vasodilating substances released during ventricular fibrillation or after defibrillation. The increase in coronary sinus oxygen content is probably a consequence of the overperfusion seen during reactive hyperemia. The increase in coronary sinus lactate and the decrease in arteriocoronary sinus lactate difference could be caused by ischemia during ventricular fibrillation or by the delivered defibrillation energy: From the present study, it is not possible to determine whether it is the preceding ventricular fibrillation with its hemodynamic consequences or the succeeding DC countershock that causes these metabolic changes. The lack of prolonged effects after repeated episodes of ventricular fibrillation and defibrillation during ICD implantation indicates that the technique used with a recovery period of 5 minutes between episodes seems to be safe as regards myocardial metabolism.
ACKNOWLEDGMENT The authors thank G. Barr, CRNA, for technical assistance, and M. Broman for assistance in preparing the manuscript. The authors are obliged to Associate Professor GSran Settergren, MD, for revising the manuscript.
REFERENCES
1. Mirowski M, Reid PR, Mower MM, et al: Termination of malignant ventricular arrhythmias with an implantable autgmatic defibrillator in human beings. N Engl J Med 303:322-324, 1980 2. Keyl C, Tassani R Kemkes B, et al: Hemodynamic ch~mgesdue to intraoperative testing of the automatic implantable cardioverter defibrillator: Implications for anesthesia management. J Cardiothorac Vasc Anesth 7:442-447, 1993 3. Hachenberg T, Hammel D, M611hoff T, et al: Cardiopulmonary effects of internal cardioverter/de fibrillation implantation. Azta Anesthesiol Scand 35:626-630, 1991 4. Runsi6 M, Bergfeldt L, Brodin L-A, et al: Left ventricular function after repeated episodes of ventricular fibrillation mtd defibrillation assessed by transoesophageal echocardiography. Eur Heart J 18:124-131, 1996 5. Franco-Cereceda A, Owall A, Settergren G, et al: Release of neuropeptide Y and noradrenaline from the human heart after aortic
occlusion during coronary artery surgery. Cardiovasc Res 24:242-246, 1990 6. Lerman B, Engelstein E: Metabolic determinants of defibrillation. Role of adenosine. Circulation 91:838-844, 1995 7. Runsi6 M, Rosenqvist M, Bergfeldt L, et al: Cardiac outflow of endotheline, neuropeptide Y and noradrenaliue in relation to byperemia in coronary sinus flow following electrical conversion of induced ventricular fibrillation in man. Eur Heart J 16:1925-1929, 1995 8. Valen G, Runsi6 M, Owall A, et al: Cardiac release of histamine after ventricular fibrillation and defibrillation during insertion of implantable cardioverter defibrillators (ICD). Inflamm Res 44:499-503, 1995 9. Tronton T, Allen J, Adgey A: Oxidative metabolism and myocardial blood flow changes after transthoracic DC countershocks in dogs. Eur Heart J 13:1431-1440, 1992 10. Trouton T, Barry J, Allen J, et al: Failure of countershock-type
50
pulses in vitro to adversely alter mitochondria oxidative phosphorylation. Ann Emerg Med 21:132-136, 1992 11. Trouton T, Allen J, Young I, et al: Altered cardiac oxygen extraction, lactate production and coronary blood flow after large dose transthoracic DC countershocks. Pace 16:1304-1309, 1993 12. Runsi6 M, Kallner A, K~illner G, et al: Myocardial injury after electrical therapy for cardiac arrhythmias assessed by Troponin T release. Am J Cardio179:1241-1244, 1997 13. Schmidt FH: Die enzymatische Bestimmung yon Glucose and Fructose. Klin Wochenschr 39:1244-1250, 1961 14. Barthelmai W, Czok R: Enzymatische Bestimmungen der Glucose in Blut, Liquor und Him. Klin Wochenschr 40:585-589, 1962 15. Jorfeldt L, Juhlin-Dannfeldt A: The influence of ethanol on human splanchnic and skeletal muscle metabolism during exercise. Scand J Clin Lab Invest 37:609-613, 1977 16. Kirk RE: Experimental Design: Procedures for the Behavioral Sciences. Belmont, CA, Brooks/Cole, 1968 17. De Piccoli B, Rigo F, Raviele A, et al: Transoesopbageal echocardiographic evaluation of the morphologic changes during ventricular fibrillation. J Am Soc Echocardiogr 9:71-78, 1996 18. Gaskell WH: The changes in the blood stream in muscles through stimulation of their nerves. JAnat 11:360-402, 1877 19. Bayliss WM: On the local reaction of the arterial wall to changes in internal pressure. J Physio128:220-231, 1902 20. Olsson RA: Myocardial reactive hyperemia. Circ Res 37:263269, 1975 21. Franco-CerecedaA, Lundberg JM: Post-occlusive reactive hyperaemia in the heart, skeletal muscle and skin of control and capsalcintreated pigs. Acta Physiol Scand 137:271-277, 1989 22. Dole WP, Montville WJ, Bishop VS: Dependency of myocardial
RUNSI(~ ET AL
reactive hyperemia on coronary artery pressure in the dog. Am J Physiol 240:H709-tt715, 1981 23. Olsson RA, Snow JA, Gentry MK: Adenosine metabolism in canine myocardial reactive hyperemia. Circ Res 42:358-362, 1979 24. Olsson RA, Gregg DE: Metabolic response during myocardial reactive hyperemia in the unanesthetized dog. Am J Physiol 208:231236, 1965 25. Olsson RA, Gregg DE: Myocardial reactive hyperemia in the unanesthetized dog. Am J Physio1208:224-230, 1965 26. Bache RJ, Cobb FR, Greenfield JC Jr: Limitations of the coronary vascular response to ischemia in the awake dog. Circ Res 35:527-535, 1974 27. Schwartz GC, McHale PA, Greenfield JC Jr: Hyperemic response of the coronary circulation to brief diastolic occlusion in the conscious dog. Circ Res 50:28-37, 1982 28. Zierler K: Theory of the use of arteriovenous concentration differences for measuring metabolism in steady and non-steady states. J Clin Invest 2111-2125, 1961 29. McKeever WP, Gregg DE, Canney PC: Oxygen uptake of the nonworking left ventricle. Circ Res 6:612-623, 1958 30. Hottenrott CE, Maloney JV, Buckberg G: Studies of the effects of ventricular fibrillation on the adequacy of regional myocardial flow. I. Electrical vs. Spontaneous fibrillation. J Thorac Cardiovasc Surg 68:615-625, 1974 31. Hottenrott CE, Buckberg G: Studies of the effects of ventricular fibrillation on the adequacy of regional myocardial flow. II. Effects of ventricular distention. J Thorac Cardiovasc Surg 68:626-633, 1974 32. Jorfeldt L. Metabolism of L(+)-lactate in human skeletal muscle during exercise. Acta Phyiol Scand 14:1-67, 1970, (suppl 338)
termination of ventricular fibrillation, coronary sinus blood flow increased to a peak value of 227 -+ 75 mL/min. Oxygen saturation, PO2, and oxygen content in the coronary sinus increased by approximately 25% shortly after each episode of ventricular fibrillation and defibrillation. The coronary sinus lactate increased and the arterio-coronary sinus lactate difference decreased shortly after each of the four episodes, but was normalized within 2 minutes. Conclusions: Repeated threshold tests during defibrillator implantation did not cause any long-lasting or cumulative metabolic effects, indicating that the described technique, with a 5-minute recovery period in between episodes, is safe as regards myocardial metabolism.
INCE THE introduction of the implantable cardioverterdefibrillator (ICD) into clinical practice, an increasing number of patients with malignant ventricular arrhythmias are now offered symptomatic treatment of an otherwise lifethreatening condition. 1 The implantation procedure, although less traumatic today, produces a temporary depression of myocardial function with the risk of further impairment in myocardial performance after repeated episodes of ventricular fibrillation and defibrillation.2-4 During the implantation procedure of an ICD, the heart is exposed to several myocardial depressant factors, in which the short-lasting ventricular fibrillation and the direct current (DC) shock delivered to the heart constitute the most probable sources of myocardM injury. Ventricular fibrillation and its effects on myocardial metabolism and the neurohumoral response have been studied ia several experimental settings.5-s So far, no study of the metabolic effects has been conducted in humans during implantation of an ICD. In animal experiments, the metabolic effects of transthoracic DC countershocks showed a large increase in coronary sinus lactate level after five defibrillator shocks. T M Other studies have shown myocardial injury indicated by the release of troponin-T and other markers of myocardial injury.l:'- The aim of this study was to evaluate the metabolic effects of repeated short episodes of ventricular fibrillation and internal defibrillation in relation to coronary artery blood flow.
6 patients, continuous coronary sinus blood flow was recorded. Clinical data for the two groups are shown in Table 1. After an overnight fast, the patients were premedicated with morphine, 0.05 to 0.1 mg/kg, and scopolamine, 2 to 4 Bg/kg.Anesthesia was induced using thiopental sodium, 3 to 5 mg/kg, and neuromuscular blockade was achieved using vecttronium, 0.1 mg/kg. Anesthesia was maintained with isoflurane in oxygen/nitrous oxide, and the patients were ventilated to achieve normocapnia. Before induction of ventricular fibrillation, the FIO2 was increased to 100%. In all 16 patients, endocardial leads (Medtronic Inc, Minneapolis, MN, or Cardiac Pacemakers Inc, St Paul, MN) were introduced through either the left cephalic or subclavian vein and combined with a subcutaneous patch, when necessary. The transvenous leads were positioned under fluoroscopic guidance, with one large coil electrode in the apex of the right ventricle and, when the Medtronic system was used, one thinner electrode was positioned in the junction of the superior vena cava and the right atrium. After an appropriate position was found, the electrodes were connected to the test equipment, and the pacing threshold and impedance as well as the slew rate and R-wave amplitude were assessed. When these values fulfilled the implantation criteria, ventricular fibrillation was induced, either by alternating current (AC) or by T-wave shock, and defibrillated with a maximal energy of 20 joules. After at least three successful tests, the electrodes were connected to the generator, which was implanted in an abdominal pocket, after which a final test with 34 joules was performed. After the induction of anesthesia, but before surgery, a thermodilution catheter (Webster Laboratories, Altadena, CA) was introduced into the coronary sinus, guided by fluoroscopy for flow measurements and blood
S
MATERIAL AND METHODS This study was approved by the Ethics Committee of the Karolinska Hospital, and the patients gave their informed consent to participate. Sixteen patients with malignant ventricular arrhythmias were studied during the implantation procedure of an ICD with regard to the effects of repeated episodes of ventricular fibrillation and defibrillation. For technical reasons, it was not possible to perform continuous measurements of coronary sinus blood flow and blood sampling in the same group of patients. In group A, consisting of 10 patients, coronary sinus blood flow and absolute and arterio-coronary sinus differences of energy substrates were intermittently assessed. In group B, consisting of
Copyright© 1998by W.B, Saunders Company KEY WORDS: ventricular fibrillation, defibrillation, implantable cardioverter-defibrillator, myocardial metabolism
From the Departments of Cardiothoracic Surgery, Cardiology, Cardiothoracic Anesthesiology, and Clinical Physiology, Karolinska Institute at Karolinska Hospital, Stockholm, Sweden. This study was supported by grants from the Swedish Heart and Lung Foundation and the Swedish Medical Research Council, Project No. 04139. Address reprint requests to Mikael Runsi5, MD, Department of Cardiothoracic Surgery, IFX Karolinska Hospital, S-171 76 Stockholm, Sweden. Copyright © 1998 by W.B. Saunders Company 1053-0770/98/1201-0008503.00/0
Journal of Cardiothoracic and Vascular Anesthesia, Vo112, No 1 (February),1998:pp 45-50
45
46
RUNSIO ET AL
Table 1. Patient Characteristics Variable/Group
Group A n - 10
Male (%)
80
Age (yr)
58 _+ 12
Failed drugs (no)
2.8 _+ 2
Group B n- 6 100 68 _+ 4 2.2 -- 1.2
Sotalol at surgery
7
4
Amiodarone at surgery
0
0
Previous heart surgery Heart disease
3
1
CAD DCM
7 1
5 0
NSHD
2
I
VT VF
4 1
4 1
VT/VF
4
1
SD
1
0
Arrhythmia
W EF Duration of operation (min) Inductions (no) VF duration (sec)
110 _+ 70 0.37 _+ 0.15
133 _+ 34 0.44 _+ 0.17
150 -- 30 6 _+ 1 16 -- 5
140 +_ 30 5 _+ 1 17 _+ 1
Abbreviations: CAD, coronary artery disease; DCM, dilated cardiomyopathy; EF, ejection fraction; NSHD, no structural heart disease; SD, sudden death; VF, ventricular fibrillation; VT, ventricular tachycardia; W, work test.
sampling. A catheter for blood sampling and blood pressure measurement was placed in the radial artery. The coronary sinus blood flow (mL/min) was determined by the retrograde thermodilution technique using a continuous infusion of saline over 20 seconds at a rate of 16.4 m L / m i n . 5 In group A patients, flow measurements were performed before the first induction of ventricular fibrillation and during the period 2 to 5 minutes after each episode of ventricular fibrillation and defibrillation, when stable hemodynamic conditions were re-established. In group B patients, the coronary sinus blood flow and the arterial blood pressure were measured continuously, and the mean value was recorded every 4 seconds, from 40 seconds before to 80 seconds after each of three episodes of ventricular fibrillation and defibrillation. The repayment/debt volume ratio was calculated by estimating the flow difference below (debt) and above (repayment) the baseline value. Blood samples were taken before induction of ventricular fibrillation, shortly after successful defibrillation, and 2 and 5 minutes after defibrillation. The 5-minute sample also served as the prefibrillation sample for the following episode, and, in each patient, 13 samples were taken from the coronary sinus and radial artery, respectively. Blood samples were obtained between 8 and 20 seconds and between 20 and 40 seconds after defibrillation for the determination of blood gas and metabolic variables, respectively. Each set of blood samples was taken in two portions: the first was drawn into a heparinized syringe and immediately placed on ice before it was analyzed for blood gas variables and calculations of base excess; oxygen saturation and oxygen were performed using standard formulas (Blood gas analyzer, model IL1312, BGM Instrumentation Laboratory Inc, Milan, Italy). The second portion, drawn directly after the blood gas portion, was immediately precipitated with ice-cold perchloric acid. After centrifugation, the protein-free extract was deep-frozen (-80°C) while awaiting analysis. The analyses were performed by microfluorimetry: D-glucose by a modification for fluorimetry of the hexokinase method, and lactate, glycerol, 3-hydroxybutyrate, and alanine as previously described.13-15 The data were analyzed in accordance with the analysis of variance (ANOVA) for repeated measures. Repeating factors were episodes with four levels and time with three levels. Because of within-subject
correlations among the repeated measurements, the p values have been adjusted according to Greenhouse-Geisser. When sphericity assumption in the repeated measures design did not hold, a multivariate analysis of variance was performed. When the overall F-ratio in the analysis of variance was significant at a 5% level, contrast (trends) between the repeated measures was calculated. The distribution for some of the variables was skewed. The data have been logtransformed to meet the requirements for an adequate ANOVA. 16 Statistical significance was defined asp < 0.05. The data are presented as mean + standard error of the mean. RESULTS
The patients in group A had a total of 6 + 1 episodes of ventricular fibrillation and defibrillations, and group B patients had 5 ± 1 episodes. The ventricular fibrillation lasted for 16 ± 5 seconds and 17 --- 1 seconds before defibrillation in groups A and B, respectively. Group A patients received 129 ± 12 joules/patient, and group B patients 149 --- 23 joules/patient. The coronary sinus blood flow in group A was 73 ± 15 mL/min at baseline and 69 ± 5 mL/min, 79 + 12 mL/min, 62 8 mL/min, and 63 ± 11 mL/min (NS), respectively, in the period 2 to 5 minutes after each of the four episodes of ventricular fibrillation and defibrillation. In group B, basal coronary sinus blood flow before the first episode of fibrillation was 93 ± 16 mL/min (Fig 1). During ventricular fibrillation, the coronary sinus blood flow decreased to 35 ± 6 mL/min (p < 0.001). Immediately after defibrillation, there was a marked increase in coronary sinus blood flow, reaching a maximum of 227 ± 75 mL/min (p < 0.001) within 20 ± 5 seconds after defibrillation. After a further 30 to 40 seconds, the coronary sinus blood flow had returned to prefibrillation levels (Fig 1). The repayment/debt volume ratio was 2.7 ± 0.4, 2.9 ± 0.3, and 3.1 ± 0.4 in three consecutive episodes of ventricular fibrillation and defibrillation, respectively. Mean arterial blood pressure decreased to 29 ± 3 mmHg during fibrillation from a basal value of 72 ± 2 mmHg (p < 0.001). It returned to the prefibrillation level of 71 ± 3 mmHg within 30 seconds following defibrillation (Fig 1). Arterial blood PO2 increased from 283 ± 34 to 417 --- 19 m m H g (p < 0.001) during the study period. Coronary sinus PO2 was stable in the prefibrillation state, but increased significantly shortly after each episode of ventricular fibrillation from a baseline level of 24 ± 1 to 28 ± 2 mmHg ( p < 0.001), 31 - 2 mmHg ( p < 0.001),29 ± 2 mmHg ( p < 0.001), and 31 ± 2 mmHg (p < 0.001), respectively. Within 2 minutes, it had returned to the prefibrillation level. The arterial oxygen saturation was 100% during the study period. The coronary sinus oxygen saturation increased significantly after each of the episodes of ventricular fibrillation and defibrillation. Before the first induction of ventricular fibrillation, the oxygen saturation in coronary sinus blood was 42% ± 2%, increasing to 50% ± 4.5% ( p < 0.001), 58% ± 3% ( p < 0 . 0 0 1 ) , 55% ± 3% ( p < 0 . 0 0 1 ) , and 56% - 5% (p < 0.001) respectively, shortly after the four episodes of ventricular fibrillation and defibrillation. The oxygen content in the coronary sinus increased significantly (p < 0.001) shortly after each of the four episodes of ventricular fibrillation and defibrillation (Fig 2). No significant changes were seen in arterial and coronary sinus PCO2, pH, standard bicarbonate, or base excess during the procedure.
MYOCARDIAL METABOLISM AND DEFIBRILLATION
47
250
80 T]TLLLLU_U_LLLLM.LLLL~
7O
T
T
[
~
_TTrTTTTTTT[TTTTTTIT
TI'T~TT
6o so
.
150
40 "~
30
'c
20
Fig 1. Mean arterial blood pressure = A and coronary sinus blood flow = ~ during three consecutive episodes of ventricular fibrillation and defibrillation in patients in group B. Mean -+ SEM.
100
I 0
r
,
20 40 60 110 100120140 0
,
,
20 40 60 80 100 120140 0
,
~
~
j
,
,
,
8
I0
20 40 60 80 100120140
Sec
The hemoglobin value was stable in both groups (luring the procedure. The mean value for group A patients was 144 4 g/L. The arterial lactate concentration was stable during the whole procedure (0.8 + 0.07 to 0.9 -+ 0.08 mmol/L) (NS). The coronary sinus lactate was stable during the prefibrillation periods. It was 0.6 +- 0.06 mmol/L at baseline and 0.7 + 0.09 mmol/L (p < 0.05), 0.7 + 0.08 mmol/L (p < 0.05) 0.7 _+ 0.06 mmol/L (p < 0.05), and 0.8 + 0.08 mmol/L (p < 0.05) after each episode of fibrillation, respectively. The arteriocoronary sinus lactate difference decreased significantly (p < 0.05) 20 to 40 seconds after each episode of ventricular fibrillation and defibrillation, but returned to baseline values 2 minutes after restoration of regular heart rhythm (Fig 3). There was a net release of lactate in four patients. There were no significant changes in alanine, glycerol, or glucose during the study period. DISCUSSION The implantation procedure of an ICD offers a unique opportunity to study the immediate effects of ventricular fibrillation and defibrillation on myocardial hemoJynamics,
neurohumoral response, and metabolism in a closed-chest model. Applying this concept, it was shown that short episodes of electrically induced ventricular fibrillation cause an immediate postdefibrillation myocardial hyperemia with a concomitant increase in coronary sinus oxygen content. These changes were associated with an increase in coronary sinus lactate and a decreased arteriocoronary sinus lactate difference. However, the observed effects were of short duration, and no signs of a cumulative metabolic derangement were observed. In the present study, the mean arterial blood pressures were stable throughout the procedure, apart from during and 40 seconds after induction of ventricular fibrillation, indicating that the hemodynamics were reasonably stable in the basic condition. In the same model, a progressive diastolic dysfunction without any concomitant deterioration in systolic function has been observed, which is in accordance with other studies. 3,4,17 Despite the reduced driving pressure during fibrillation, there was a persistent, although decreased, myocardial perfusion. This reduction is in accordance with results from previous studies. 4,7-8 After defibrillation, a marked hyperemia was observed. Reactive hyperemia is a well-kalown physiologic reac0.30
14.0
0.25
12.0
t
10,0
E x O
t
~
0.20
8,0
0.15
6.0 4.0
0.10 2.0
J
0.05
0.0 BI
A0
Episode 1
A2
B2
A0
Episode 2
A2
B3
A0
Episode 3
A2
B4
AC
A2
A5
Episode 4
Fig 2. Oxygen content in coronary sinus during four consecutive episodes of ventricular fibrillation and defibrillation in group A patients. Mean + SEM. Abbreviations: B1-B4, before eph,;ode 1 to 4; A0, shortly after defibrillation; A2, 2 minutes after defibrillation; and A5, 5 minutes after defibrillation.
0.00
A
A2
A5
Fig 3. Arteriocoronary sinus lactate difference during four consecutive episodes of ventricular fibrillation and defibrillation in group A patients. Mean -+ SEM. Abbreviations: B, before; A, after defibrillation; A2, 2 minutes after defibrillation; A5, 5 minutes after defibrillation; r~, Episode 1; II, Episode 2; [], Episode 3; D , Episode 4.
48
A
RUNSIO ET AL
B
6oo
600 o
.E
O
E •~. 500
500
O
O
0
o 400
o
8
o
o
.El
400
o
o
o
c
'~ 200
~o
o
0
o
Oo
° i 0 -°
oll
0 8e u 8°oO ~e ~o o g ^ l l
O0
100
2 o
p-II-dl'-1F8
L)
0
I
8~
8
o
o
i
I
I
I
20
40
60
80
Mean arterial blood pressure
(ram Hg}
0
8
o 300
'~ 200
0
O O
300
O
o o°
lo0
°o o - e ~ -
@
O
°i
o 0
0
i
I
I
i
20
40
60
80
Mean arterial blood pressure (mm Hg}
Fig 4. (A) Mean arterial blood pressure versus coronary sinus blood flow before induction of, and during, ventricular fibrillation in group B patients. The line indicates correlation using the least square method. Arrow indicates direction of pressure and coronary sinus blood flow development. (B) Mean arterial blood pressure versus coronary sinus blood flow immediately after defibrillation and during reactive hyperemia in group B patients. Interrupted line indicates correlation using the least square method. Arrow indicates direction of pressure and coronary sinus blood flow development.
tion after reperfusion after regional impairment of arterial blood flow.18,19This phenomenon is considered to be caused by both physical and metabolic factors, and has been observed in several organs, including the heart. 2°-22Figure 4 is a scatterplot of coronary sinus blood flow versus mean arterial blood pressure during fibrillation with successively decreasing mean arterial blood pressure (Fig 4A) and during reperfusion (Fig 4B). During fibrillation, coronary sinus blood flow decreased approximately in parallel with the mean arterial blood pressure. This could indicate that the coronary vascular resistance was essentially unchanged. During reperfusion, approximately the same relationship occurred, initially at low mean arterial blood pressures. 22 After this, however, the coronary sinus blood flow was higher than predicted from the mean arterial blood pressure, indicating a decrease in the coronary vascular resistance probably because of the release of vasoactive agents, a3 Adenosine release may be one plausible mechanism responsible for the postfibrillation/defibrillationhyperemia seen in these patients. The release of adenosine, in association with the reactive hyperemic response seen after occlusion of a coronary artery or after ventricular fibrillation, has been observed by others. 6,23In the present study, there was a hypoperfusion of 17 -+ 1 second, approximately the same duration as in the study by Olsson et al,2° with a considerable hyperemic response shortly after defibrillation. Other investigators have studied the volume repayment/debt ratio and found that the repayment volume always exceeds the debt volume, which is in accordance with the authors' current results. 18,z.26 The repayment/debt ratio in the present study was 3:1, as has been described by others in occlusion studies of the same duration. This overperfusion could indicate a temporary uncoupling between oxygen supply and demand. The very high efficiency of the compensatory mechanisms involved in the regulation of myocardial perfusion is further illustrated in a study by Schwartz et al,27 in which occlusion episodes as brief as 200 milliseconds were enough to create an immediate hyperemic response during the next heartbeat.
In the present study, the coronary sinus oxygen saturation increased by 25% shortly after the fibrillation/defibrillation episodes. Coronary sinus oxygen saturation could not be measured continuously during ventricular fibrillation. Therefore, the myocardial oxygen debt and repayment could not be calculated. Considering, however, that the repayment/debt ratio of coronary sinus blood flow was high (3:1) when compared with the moderate increase in the coronary sinus oxygen saturation, it seems as if an oxygen debt was not only repaid, but, to some extent, overcompensated for. This reaction is probably a consequence of the hyperemia alone. An alternative cause of the increase in coronary sinus oxygen content could be a depressed mitochondrial oxidative phosphorylation in response to the delivered DC countershocks, as described by Trouton et 1tl. 9-11 The response to the combined arrhythmic and electrical trauma is Complex, and several interacting factors probably contribute to the observed increase in coronary sinus oxygen content.29 Coronary sinus lactate increased shortly after defibrillation, with a concomitant reduction in the arteriocoronary sinus lactate difference seen particularly after the fourth episode. However, measurements in association with ventricular fibrillation and defibrillation are generally made under non-steadystate conditions, complicating the interpretation of the results. 28 To reduce this problem, a comparison between arteriocoronary sinus differences of lactate with that of oxygen was performed (Fig 5). Provided that the compared substances have a similar distribution volume and transit time through the tissue, a similar response to the intervention could be predicted. 28 If this could be assumed for lactate and oxygen, it further supports the ideas put forward in this study, indicating a decreased lactate uptake seen shortly after the episodes of DC cardioversion, most prominent after episodes 1 and 4 (Fig 5). An increase in coronary sinus lactate during these non-steady-state conditions may indicate ischemia, although a net release of lactate is the only reliable sign of ischemia. Coronary sinus lactate increased after veutricular fibrillation, concomitant with a pronounced
MYOCARDIAL METABOLISM AND DEFIBRILLATION
49
0.06 0.05
ttttt t
0.04 0.03 0.02 0.01 0.00 -0.01 -0.02
B1
A0
Episode 1
A2
B2
AO
Episode 2
A2
B3
A0
Episode 3
A2
B4
A0
A2
A5
Episode 4
Fig 5. The arteriocoronary sinus lactate versus oxygen difference in group A patients. Mean _+ SEM. Abbreviations: B1-B4, before episode 1 to 4; A0, immediately after defibrillation; A2, 2 minutes after defibrillation; and A5, 5 minutes after defibrillation.
increase in coronary sinus blood flow, but without corresponding increase in oxygen demand. Lactate concentration may have increased because of ischemia, but may also have been altered by the marked increase in myocardial blood flow and the corresponding increase in substrate (lactate) delivery. Determination of lactate during reactive hyperemia may, therefore, be a blunt tool for the detection of myocardial ischemia. The coronary artery pressure was severely depressed for 20 to 30 seconds, which could be one possible cause of the effects on lactate metabolism seen in these patients. This is in accordance with studies by Olsson et al,u,25 in which a hypoperfusion of 6 seconds was enough for lactate production to appear. As described by Trouton et al, 9-n the electrical energy delivered in sinus rhythm could cause a significant lactate release. This seemed to be a dose-response effect because this effect was caused by five shocks, but not by two. Thus, both the low coronary artery pressure cau:;ed by the arrhythmia and the defibrillation energy could be of importance for the changes seen in myocardial metabolism, as indicated by the decreased arteriocoronary sinus lactate difference: after ICD threshold tests. This study was performed on a limited number of patients
with various underlying diseases and ventricular arrhythmias, which reflects the clinical situation. However, the data are in accordance with previous animal studies. 9-11 Another limitation is that the experimental setup did not allow simultaneous measurements of blood flow and blood sampling from the coronary sinus, which makes all attempts to calculate ~:he absolute release or uptake of different substances precarious. A further limitation is that the sampling of metabolites and blood gas components in coronary sinus would only give a net result of the release of the measured substance. If regional productmn and release of a substance, as described for lactate, could be suspected, this method has obvious disadvantages, raising the possibility of failing to detect a significant release of clinical importance. 3°-32 The measurements were made shortly after defibrillation, during non-steady-state conditions, further cemplicating the analysis of the results as previously discussed. The decrease in coronary sinus blood flow after induction of ventricular fibrillation is probably an effect of decreased driving pressure over the coronary vascular bed, whereas the postdefibrillation hyperemia seen in all patients after defibrillation might be a result of vasodilating substances released during ventricular fibrillation or after defibrillation. The increase in coronary sinus oxygen content is probably a consequence of the overperfusion seen during reactive hyperemia. The increase in coronary sinus lactate and the decrease in arteriocoronary sinus lactate difference could be caused by ischemia during ventricular fibrillation or by the delivered defibrillation energy: From the present study, it is not possible to determine whether it is the preceding ventricular fibrillation with its hemodynamic consequences or the succeeding DC countershock that causes these metabolic changes. The lack of prolonged effects after repeated episodes of ventricular fibrillation and defibrillation during ICD implantation indicates that the technique used with a recovery period of 5 minutes between episodes seems to be safe as regards myocardial metabolism.
ACKNOWLEDGMENT The authors thank G. Barr, CRNA, for technical assistance, and M. Broman for assistance in preparing the manuscript. The authors are obliged to Associate Professor GSran Settergren, MD, for revising the manuscript.
REFERENCES
1. Mirowski M, Reid PR, Mower MM, et al: Termination of malignant ventricular arrhythmias with an implantable autgmatic defibrillator in human beings. N Engl J Med 303:322-324, 1980 2. Keyl C, Tassani R Kemkes B, et al: Hemodynamic ch~mgesdue to intraoperative testing of the automatic implantable cardioverter defibrillator: Implications for anesthesia management. J Cardiothorac Vasc Anesth 7:442-447, 1993 3. Hachenberg T, Hammel D, M611hoff T, et al: Cardiopulmonary effects of internal cardioverter/de fibrillation implantation. Azta Anesthesiol Scand 35:626-630, 1991 4. Runsi6 M, Bergfeldt L, Brodin L-A, et al: Left ventricular function after repeated episodes of ventricular fibrillation mtd defibrillation assessed by transoesophageal echocardiography. Eur Heart J 18:124-131, 1996 5. Franco-Cereceda A, Owall A, Settergren G, et al: Release of neuropeptide Y and noradrenaline from the human heart after aortic
occlusion during coronary artery surgery. Cardiovasc Res 24:242-246, 1990 6. Lerman B, Engelstein E: Metabolic determinants of defibrillation. Role of adenosine. Circulation 91:838-844, 1995 7. Runsi6 M, Rosenqvist M, Bergfeldt L, et al: Cardiac outflow of endotheline, neuropeptide Y and noradrenaliue in relation to byperemia in coronary sinus flow following electrical conversion of induced ventricular fibrillation in man. Eur Heart J 16:1925-1929, 1995 8. Valen G, Runsi6 M, Owall A, et al: Cardiac release of histamine after ventricular fibrillation and defibrillation during insertion of implantable cardioverter defibrillators (ICD). Inflamm Res 44:499-503, 1995 9. Tronton T, Allen J, Adgey A: Oxidative metabolism and myocardial blood flow changes after transthoracic DC countershocks in dogs. Eur Heart J 13:1431-1440, 1992 10. Trouton T, Barry J, Allen J, et al: Failure of countershock-type
50
pulses in vitro to adversely alter mitochondria oxidative phosphorylation. Ann Emerg Med 21:132-136, 1992 11. Trouton T, Allen J, Young I, et al: Altered cardiac oxygen extraction, lactate production and coronary blood flow after large dose transthoracic DC countershocks. Pace 16:1304-1309, 1993 12. Runsi6 M, Kallner A, K~illner G, et al: Myocardial injury after electrical therapy for cardiac arrhythmias assessed by Troponin T release. Am J Cardio179:1241-1244, 1997 13. Schmidt FH: Die enzymatische Bestimmung yon Glucose and Fructose. Klin Wochenschr 39:1244-1250, 1961 14. Barthelmai W, Czok R: Enzymatische Bestimmungen der Glucose in Blut, Liquor und Him. Klin Wochenschr 40:585-589, 1962 15. Jorfeldt L, Juhlin-Dannfeldt A: The influence of ethanol on human splanchnic and skeletal muscle metabolism during exercise. Scand J Clin Lab Invest 37:609-613, 1977 16. Kirk RE: Experimental Design: Procedures for the Behavioral Sciences. Belmont, CA, Brooks/Cole, 1968 17. De Piccoli B, Rigo F, Raviele A, et al: Transoesopbageal echocardiographic evaluation of the morphologic changes during ventricular fibrillation. J Am Soc Echocardiogr 9:71-78, 1996 18. Gaskell WH: The changes in the blood stream in muscles through stimulation of their nerves. JAnat 11:360-402, 1877 19. Bayliss WM: On the local reaction of the arterial wall to changes in internal pressure. J Physio128:220-231, 1902 20. Olsson RA: Myocardial reactive hyperemia. Circ Res 37:263269, 1975 21. Franco-CerecedaA, Lundberg JM: Post-occlusive reactive hyperaemia in the heart, skeletal muscle and skin of control and capsalcintreated pigs. Acta Physiol Scand 137:271-277, 1989 22. Dole WP, Montville WJ, Bishop VS: Dependency of myocardial
RUNSI(~ ET AL
reactive hyperemia on coronary artery pressure in the dog. Am J Physiol 240:H709-tt715, 1981 23. Olsson RA, Snow JA, Gentry MK: Adenosine metabolism in canine myocardial reactive hyperemia. Circ Res 42:358-362, 1979 24. Olsson RA, Gregg DE: Metabolic response during myocardial reactive hyperemia in the unanesthetized dog. Am J Physiol 208:231236, 1965 25. Olsson RA, Gregg DE: Myocardial reactive hyperemia in the unanesthetized dog. Am J Physio1208:224-230, 1965 26. Bache RJ, Cobb FR, Greenfield JC Jr: Limitations of the coronary vascular response to ischemia in the awake dog. Circ Res 35:527-535, 1974 27. Schwartz GC, McHale PA, Greenfield JC Jr: Hyperemic response of the coronary circulation to brief diastolic occlusion in the conscious dog. Circ Res 50:28-37, 1982 28. Zierler K: Theory of the use of arteriovenous concentration differences for measuring metabolism in steady and non-steady states. J Clin Invest 2111-2125, 1961 29. McKeever WP, Gregg DE, Canney PC: Oxygen uptake of the nonworking left ventricle. Circ Res 6:612-623, 1958 30. Hottenrott CE, Maloney JV, Buckberg G: Studies of the effects of ventricular fibrillation on the adequacy of regional myocardial flow. I. Electrical vs. Spontaneous fibrillation. J Thorac Cardiovasc Surg 68:615-625, 1974 31. Hottenrott CE, Buckberg G: Studies of the effects of ventricular fibrillation on the adequacy of regional myocardial flow. II. Effects of ventricular distention. J Thorac Cardiovasc Surg 68:626-633, 1974 32. Jorfeldt L. Metabolism of L(+)-lactate in human skeletal muscle during exercise. Acta Phyiol Scand 14:1-67, 1970, (suppl 338)